Current protocols for ameliorating renal allograft rejection tend to generate an all-encompassing immunosuppression and often have undesirable side effects both related and unrelated to the immunosuppressed state. This proposal constitutes a basic study for designing novel immunotherapeutic agents that would much more finely tune immune intervention through the modulation of lymphocyte subsets. It is now clear that expression of a combination of define T cell antigens is associated with function. Accordingly, in addition to the pan-T cell markers (e.g. CD2, CD3, CD5, CD6) subsets of the T cell population have been defined through CD4 (helper/inducer T ells) and CD8 (suppressor/cytotoxic T cells). The CD4 subset may be further divided firstly, into helper-inducer (memory) cells through the expression of CD45RO, CD29, and upon activation, CD26 and secondly, into suppressor- inducer (naive) cells through the expression of CD45RA. Meanwhile, the CD8 population may be split into suppressor (CD8+S6F1-) and cytotoxic populations (CD8+S6F1+). We will produce new monoclonal antibodies against epitopes upon the T cell antigens described above, with a view not only for selecting by affinity, but also with potential as an immunotoxin in mind. We will thus look for epitopes which upon binding and crosslinking encourage internalisation. We will use the property of internalisation only after crosslinking to develop a panel if bifunctional antibodies that may simultaneously target two T cell antigens and deliver a toxin moiety upon binding bivalently. Immunotoxins produced from such antibodies will then be tested extensively in vitro for their ability to selectively deplete helper- inducer cells and cytotoxic cells while leaving suppressor-inducer cells and suppressor cells intact. Furthermore, we will use unconjugated bifunctional antibody to crosslink and attempt to activate suppressor- inducer cells (CD4+CD45RA) to further enhance suppression. We believe that the selective reduction in the helper T cell population while leaving the suppressor cell pathway intact will provide a much more delicate handle than is currently available for modulating T cell function in vivo and, as such, we will undertake these experiments with the intention of eventually testing the developed reagents in a non- human primate renal allograft model. One of the major complications of introducing therapeutic antibodies in vivo, however, is the initiation of a recipient anti-antibody response. Consequently, we will also undertake the production, from antibodies selected as efficient in vitro, of chimeric or fully humanised monoclonal and bifunctional antibodies. This provides the added bonus of enabling us to move antibody production into high expression vectors, which will be essential if we wish to later produce large amounts of immunotoxin for testing in vivo.